Table 4.
Performance of GBM using Poisson distribution in the prediction of recovered and active cases of COVID-19 in different states of India
| Performance parameter | State | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Maharashtra | Gujarat | Kerala | AP | Haryana | WB | Telangana | Rajasthan | UP | MP | |
| Recovered cases | ||||||||||
| MSE | 2084.27 | 481.27 | 533.36 | 1248.65 | 280.65 | 64.68 | 470.60 | 533.36 | 870.35 | 2046.96 |
| RMSE | 45.65 | 21.94 | 23.09 | 35.34 | 16.75 | 8.04 | 21.69 | 23.09 | 29.50 | 45.24 |
| MAE | 21.45 | 14.14 | 13.45 | 22.38 | 13.20 | 6.27 | 14.11 | 13.45 | 14.51 | 22.42 |
| MRD | −1970.83 | −302.02 | −821.56 | 1248.65 | −324.46 | −129.59 | −606.42 | −821.58 | −274.23 | −276.50 |
| R2 | 0.95 | 0.77 | 0.98 | 0.75 | 0.84 | 0.89 | 0.91 | 0.98 | 0.43 | 0.21 |
| Active cases | ||||||||||
| MSE | 85,646.60 | 67,422.76 | 55,572.56 | 9931.23 | 904.03 | 6042.37 | 61,793.21 | 88,292.87 | 29,361.50 | 40,995.38 |
| RMSE | 292.65 | 259.66 | 235.74 | 99.66 | 30.07 | 77.73 | 248.58 | 297.14 | 171.35 | 202.47 |
| MAE | 126.34 | 154.05 | 117.85 | 73.29 | 25.95 | 50.03 | 139.48 | 181.86 | 96.00 | 96.04 |
| MRD | −21,684.12 | −6515.20 | −10,037.81 | 9931.23 | −677.64 | −902.82 | −3738.07 | −6597.99 | −4589.25 | −5944.38 |
| R2 | 0.95 | 0.83 | 0.84 | 0.75 | 0.64 | 0.29 | 0.19 | 0.61 | 0.73 | 0.82 |
Note: R2 (coefficient of determination), MSE (mean square error), RMSE (root mean square error), MAE (mean average error), MRD (mean residual deviance), AP (Andhra Pradesh), WB (West Bengal), UP (Uttar Pradesh), and MP (Madhya Pradesh)